JP3039607U - Unit filter for pyrotechnic inflator - Google Patents

Abstract

(57) [Abstract] [Problem] For a pyrotechnic airbag restraint module,
Provide a unitary monolithic filter that is simple and inexpensive to manufacture and easy to install. The filter has a first layer having a high porosity and a second layer having a low porosity. The first layer 48 is preferably made from inexpensive ceramic grit 52 that is sintered and agglomerated, or made from a hard reticulated ceramic. The second layer 50 is the metal particles 5
Made from 6, increases strength due to brittle ceramics. The metal particles 56 form the second layer 50 in place,
It may then be sintered or made separately to form the first layer 4
8 may be fixed in place, or the first layer 4
8 may also be applied by plasma arc growth.

Description

[Detailed description of the invention]

[0001]

[Technical field to which the invention belongs]

 The present invention relates generally to pyrotechnic airbag restraint modules. More particularly, the present invention relates to an improved unitary monolithic filter for such a module.

[0002]

[Prior art]

 Pyrotechnic airbag suppression modules are well known and include an inflator tethered to a cushion. The cushion is inflated by the generation of gas by the material stored in the inflator housing. However, with the most commonly used gas generant materials, undesired amounts of solids are produced during gas evolution, where they are reduced to acceptable levels within the housing. A filter is provided. The filter also acts as a heat sink to cool gases that can be extremely hot.

Although various shapes and sizes of filters are known, two of the most common filter shapes are referred to herein as disks and tubes. Disc filters are circular flat elements, commonly used in axial flow inflators. The second type of tubular filter takes the form of a single hollow tube. Both of these filters are typically made of multiple layers of various filter materials such as wire mesh and ceramic paper.

Such multilayer filters are difficult and expensive to manufacture and install, especially in the case of tubular filters. In order to avoid these disadvantages, various attempts have been made to provide airbag filters in the form of unitary elements that can be easily produced in large quantities and easily attached to the housing. Unit filters, of course, must withstand the intense heat and pressure associated with pyrotechnic airbags. To withstand heat, it has been proposed to make filters from agglomerated ceramic particles or grits. An example of this is shown in JP-A-50-48797 published on May 1, 1975.

These ceramic filters are very heat resistant and also very resistant to compression. However, they are quite weak in tension. As a result, ceramics are mostly effective for disk filters that are placed under compression, but are not suitable for tubular filters that generate large hoop stresses.

Various other materials have been proposed, such as reticulated carbon foam with a metal coating, as shown in US Pat. No. 5,372,380 to Duffy et al. There is. Although such materials can provide the desired heat and pressure resistance, they are expensive to manufacture.

[0007]

[Problems to be solved by the device]

 It is an object of the present invention to provide an airbag filter that is simple and inexpensive to manufacture and easy to install.

Another object of the present invention is to provide a filter as described above, which withstands the heat and pressure of air bag actuation while providing the desired filter action.

Yet another object of the present invention is to provide a filter material of sufficient strength that it may be used as an outer housing for an inflator.

[0010]

[Means for Solving the Problems]

 These and other objects are accomplished by a reinforced ceramic unitary filter for airbags. These filters have a first layer of high porosity and a second layer of low porosity. This first layer is preferably made from inexpensive ceramic granules that are sintered and agglomerated, or made from a hard reticulated ceramic. The second layer is made of metal particles and increases strength due to the brittle ceramic. The metal particles may be in place in a second layer and then sintered. Alternatively, the second layer may be made separately and fixed in place on the first layer. As a third alternative, the second layer may be applied by plasma arc growth on top of the first layer. The metal particles increase the strength of the filter and also lead to a finer filter layer. The metal layer can also preload the ceramic layer to compress the ceramic and increase resistance to fracture.

[0011]

[Embodiment of the invention]

 The above objects and features of the present invention will be described in detail with reference to the drawings. Like reference numerals in the drawings denote like elements.

The present invention is primarily concerned with components and internal structure for unitary filters for airbags and other pyrotechnic inflators. Therefore, in order to better describe the invention, some of the more common final shapes for filters and inflators will be described first.

Referring to FIG. 1, a disc filter according to the present invention is generally designated by reference numeral 10. The filter 10 is approximately disk-shaped and has an upper surface 12 and a lower surface 14, and a peripheral side wall 16. Disc filters are typically used in driver side airbag inflators having a cylindrical shape, shown generally as 18 in FIG.

The inflator 18 includes a top wall 20 and a bottom wall 22, which are joined together by a peripheral side wall 24. Within these walls is a tubular inner wall 26 that defines an inflator cavity 28 for containing a first pyrotechnic material. A set of gas outlets 30 extends through the inner wall 26 to an annular generant chamber 32 defined between the inner wall and the peripheral side wall 24. This generant chamber contains a second pyrotechnic material 34, typically referred to as the generant, which produces most of the gas used to inflate the cushion (not shown) of the airbag. There is.

The filter 10 is mounted in the generant chamber 32 between the upper wall 20 and the generant 34. The top wall 20 includes a set of outlets 36 through which the gas generated by the generant can exit the inflator. As will be appreciated, this gas must pass through the disc filter 10 before exiting exit 36. Thus, the filter can remove particulate matter from the gas and cool the gas.

A slightly modified version of this configuration is shown in FIG. 4, where like reference numbers represent like elements. In this driver side inflator, the only difference is the location of the outlet 36, which penetrates the side wall 24. Nevertheless, these outlets are arranged so that the gas must pass through the filter before leaving these outlets. A blocking wall 38 may be provided between the generant and the filter adjacent the outlet to ensure that the particulates pass through the filter sufficiently to cool and cool. This barrier wall directs the gas radially to the inner portion of the filter, passing the gas through a portion of the filter sufficient to cause cooling and filtering. Further, the filter in this configuration may have its upper surface 12 sealed to prevent gas from passing between the filter and the upper wall 20 of the inflator. This sealing can be done with known sealant compounds or with layers of ceramic slurries made from alumina, silica or zirconia.

A second conventional filter type tubular filter is shown in FIG. 5, which is designated by reference numeral 40. The tubular filter employs a tubular cross-sectional shape, with the sidewall 42 ending at a longitudinal (longitudinal) end 44. This tubular filter is commonly used in both driver and passenger side inflators, with appropriate changes in tube size and dimensions.

A driver side inflator is shown in FIG. 6 and includes many of the same elements as in the inflator of FIGS. 2 and 4. The difference here is that the tubular filter wraps around the inner surface of the side wall 24 so that it lies between the generator 34 and the side wall 24 containing the outlet. A typical passenger side inflator is similar in shape to the driver side inflator of FIG. 6 with some differences. First, passenger inflators are typically longer. Also, there is no rigid inner wall 26 surrounding the initiator chamber 28, instead there is simply a foil wrapper.

There are of course many variations within the scope of prior art inflators. One variant is of particular use for the unitary filter according to the invention, which is the filter housing, although not all variants can be listed here because there are too many.

As shown in FIG. 7 for the passenger side inflator, some (or all) of the housing (top wall, bottom wall, sidewalls) can be made from a unitary filter material, thus providing a functional filter. And the functionality of the structural housing can be combined into a single component. In the example shown, the typical sidewalls of the inflator are replaced by filter sidewalls 46 made from tubular filters. The housing is more complete with top and bottom walls (commonly referred to as end caps for passenger inflators) that can be secured to the filter sidewalls in various known ways. With the appropriate modification, this same principle can be used in the driver side inflator. Thus, the filter according to the present invention may be used within the inflator housing or may form at least a portion of the housing.

Thus, while showing various examples of final filter morphology or shape, and various examples of their use in various inflator types, the materials identified below It is to be understood that can be any other shape suitable for pyrotechnic inflators and still remain within the scope of the present invention.

Next, the specific components of the filter of the present invention will be examined in more detail. There are two main layers for each of the filters according to the invention. These are typically the first layer 48, upstream of the filter or on the side entering the filter, and the second layer 50, typically on the side downstream of or out of the filter. Consists of. Although these layers may be further subdivided into sub-layers, the general form of the first and second layers remains the same for each aspect of the invention.

Keeping in mind two basic forms of filter, disc and tubular, the most common configuration of two layers is a disc and a tube. For example, for the filter of Figure 1, the overall shape is a disc and the gas flows along the longitudinal axis. Thus, the first and second layers are both in the shape of a disc, with the first layer at the bottom, next to the generant, and the second layer at the top. , Next to the exit.

With respect to the filter of FIG. 3, the overall shape is still a disc, but the gas begins to flow longitudinally, as in the case of the inflator of FIG. 4, in the radial direction. Get out. Thus, the first layer is in the shape of a central disc, while the second layer is tubular and is located on the side wall of the filter. Finally, referring to the tubular filters of Figures 5 and 7, the gas flow is completely semi-radial (radial) and the first layer is in the form of a radially inner tube and the second layer is Form an outer tube in the radial direction.

As noted above, and as is common in the art, the first layer generally has a large porosity and serves primarily to cool the gas. The second layer has a lower porosity and serves primarily to remove particles. Of course, the first layer has some filtering action, and the second layer has some cooling action.

The first layer could be made of various materials. Although there are various material possibilities, it is preferable to use a ceramic because of its low price. The ceramic may be in the form of bonded ceramic particles 52 (FIG. 8) or in the form of reticulated foam 54 (FIG. 9).

In order to make the first layer from bound particles 52, the loose agglomerates of these particles, which are typically mixed with a vitreous binder, are combined into a filter shape. (That is, a disc or the like). This mixture, with a consistency similar to moist sand, is shaped to at least approximate the final desired shape. This can be done by extruding the mixture, isostatically pressing it or pressing it in a mold to make preforms or greens. The untreated product is then processed (eg, by simple drying, calcination and / or sintering) to produce a solid, solid unit filter.

The particle size may vary depending on the desired porosity of the first layer. For most applications, particles in the range of about 0.053 inches (1.35 mm, or 10 mesh) to 0.012 inches (0.30 mm, or 50 mesh) are acceptable. Various ceramic materials such as silicon carbide and alumina can be used as these particles.

For reticulated ceramic foams, a sacrificial foam, typically an open cell polyurethane, is dipped into a ceramic slurry, such as a slurry formed from silica, zirconia or alumina, or otherwise sacrificial foam. Apply. The ceramic material is dried and then the ceramic coated sacrificial foam is heated to burn out the polyurethane sacrificial foam, leaving a ceramic material in the form of a reticulated foam ceramic 54. Ceramic foam 54 having porosity in the range of 10 to 70 pores per inch (2.54 cm) is typically acceptable for the first layer.

For both materials, the ceramic offers good thermal and filter properties. However, as is known in the art, these ceramics are not tensile strong. Thus, the force of gassing that puts the filter under tension can destroy the ceramic filter material and reduce its effectiveness. This is especially true for tubular filters that are exposed to hoop stress due to the evolution of gas.

To solve this problem, the second layer 50 of the present invention adds strength to the filter. This second layer 50 is made of metal and, in addition to filtering and cooling, provides a strong layer that serves to reinforce the brittle first layer.

The second layer is an aggregate of metal particles 56. The particle size may vary depending on the desired porosity of the first layer. For most applications, particles in the range of about 0.053 inches (1.35 mm, or 10 mesh) to 0.012 inches (0.03 mm, or 50 mesh) are acceptable. Various materials can be used as the particles 56 so long as the melting point is above the temperature that one would expect to encounter in the inflator. Some suitable metals are stainless steel, carbon steel, aluminum and nickel.

There are three different methods used to make the second layer of metal particles. The first two methods involve shaping the mass of particles into the desired shape and then sintering the particles. The difference between these first two methods is whether the particles are molded on top of the first ceramic layer into their shape or separated from the first ceramic layer into their shape. . A third method of making the second layer of metal is plasma arc growth.

Not all methods can be used for all filter configurations. Specifically, if the second layer is made as a flat disc as in Figure 1, the untreated product is by pressing a mass of particles on top of the first layer before sintering. Alternatively, it must be made directly by plasmatak growth. However, if the second layer is in the form of a tube, any of the three methods may be used. The reason for this is explained more fully below.

The first method of making the second layer of metal is to press the particles of the second layer of metal to the first layer. As mentioned above, the first layer may be a ceramic foam or a ceramic agglomerate. Generally, for both materials, the first layer is placed in a suitable mold. Then, a mass of metal particles is also put in this mold and compressed. This compaction brings the metal particles into their shape as green. The particles then retain their shape, either by using a binder or simply due to the compressive forces if they are sufficiently large. Besides this, this compaction can fix the raw product to the first layer.

Referring to FIG. 8, an enlarged view of the sintered ceramic agglomerates 52 of the first layer 48 and the metal particles 56 of the second layer 50 is shown. As will be appreciated, if the dimensions of the ceramic particles are sufficiently larger than the dimensions of the metal particles, the particles 56 can intercalate into the interstices of the first layer. This contact between the ceramic and the metal serves to hold the second layer against the first layer, especially when a binder is used. Moreover, allowing the metal to enter the serpentine gap can result in a number of interfering bonds, which further helps hold the second layer to the first layer. This is especially true after sintering.

Similarly, FIG. 9 shows the metal particles of the first layer 48 and the second layer 50 of reticulated carbon foam. Again, if the porosity of the ceramic is sufficiently larger than the metal particles, the particles 56 can be forced into these pores. This results in contact and interlaced joints, as in the example above.

If the green article is made directly on the first layer of ceramic, this combination of first and second layers is sintered. This is done at a temperature more appropriate for the metal than for the preformed ceramic first layer. This sintering process serves to firmly secure the particles to each other to complete the second layer and also to further secure the metal layer to the ceramic first layer.

Next, a second method of forming the second layer of metal will be described. This method is believed to be the same whether the first layer is a ceramic foam or a ceramic agglomerate. This method seems to be suitable only for tubular ones, not for flat layers, as explained below.

In this method, as before, the first layer 48 is made and sintered. The mass of metal particles is then pressed in a mold and bonded to form a tubular second layer of untreated product. This pressing is done without the first layer, so that the untreated article contains only the second layer of metal. During this molding, the inner diameter of the second layer 50 is dimensioned to fit the outer diameter of the disk or tubular first layer 48, similar to a press fit. The green article of this second layer is then sintered.

Thereafter (as soon as possible after the second layer has been sintered), the tubular second layer 50 is slid over the outer diameter of the first layer 48. The second layer metal has a higher coefficient of thermal expansion than the first layer of the ceramic, so that the second layer shrinks more than the first layer during cooling. Thus, this creates a binding fit that secures the second layer to the first layer. Furthermore, this greater shrinkage of the second layer 50 puts the first layer 48 under compression. This serves to preload the first layer of ceramic and increase the tensile load it can withstand.

A somewhat similar effect can also be obtained by sintering a second layer of untreated article on a first layer of ceramic according to the first method. Specifically, cooling causes the tubular second layer to shrink over the inner first layer and place it under compression. This effect can be enhanced by maintaining the pressure on the second layer 50 during the sintering process.

As will be appreciated, in the second method, this contracting effect is utilized to hold the tubular second layer tightly on the disc or tubular first layer. It should also be clear that this contraction cannot be used to secure the flat second layer of FIG. Specifically, although the interlacing that occurs during the initial pressing could somewhat increase the interlacing and compression of the first layer, this is the minimum, and the two layers together It doesn't help to maintain enough.

In addition to the above two methods, there is a third method of providing a second layer of metal particles. This is a plasma arc spray, also known as spray growth. During spray deposition, a powder, typically a metal powder, is introduced into the torch arc. The particles are propelled toward the growth surface by the torch and simultaneously heated. This heating is sufficient to bond the metal particles to the growth surface (and to each other), but does not really melt the particles. Thus, the growth surface receives a coating of fine particles.

The first ceramic layer used in the present invention is first shaped and sintered. A thin coating of metal particles is then applied over the first layer using standard plasma arc spray equipment, thus forming a second metal layer. As is known in plasma arc growth, the filter may be rotated, reciprocated, or otherwise moved during the growth process to obtain a uniform coating.

Although the equipment and methods are standard, there are limitations on the powders that can be grown (deposited). Specifically, the powder is a metal with a melting point high enough to withstand the temperatures that the final metal layer develops during operation of the inflator. Typical metals that can be used are stainless steel, carbon steel, aluminum and nickel.

The particle size used may be the same as the powder to be sintered, ie in the range of about 50-200 μm. These particles were applied by this method in a random fashion, essentially the same as shown in Figures 8 and 9, but with less mixing as much as possible. It It may be necessary to make several arcs to the first layer to obtain sufficient thickness. The final thickness depends on various factors including the porosity of the first ceramic layer, the strength of the first ceramic layer, the particle size of the metal, and the desired final porosity. Generally, the particles are applied in a thickness sufficient to provide the necessary strength to support the first layer and to provide the final porosity desired for the finished filter.

For each of the above methods (ie, the sintered powder metal layer or the plasma arc metal layer), the final (ie, inner and outer layers combined) filter was a 0.5 inch (12.7 mm) water column. It is preferred to have a porosity of about ^{2 to} 15 CFM per 1 ft ² (0.093 m ² ) of filter surface area at pressure. Of course, this can be obtained with various combinations of the porosity of the first layer and the porosity of the second layer. The specific porosity of any individual layer (or secondary layer) is not critical as long as sufficient cooling is achieved, sufficient filtering action is obtained, and porosity in the 2-15 C FM range is obtained. .

Although the above configurations are acceptable, it is possible to implement the invention with modifications to the first layer 48 of the ceramic aggregate 52. Possible modifications can make the first layer 48 stronger and / or have better gas diffusion. Nevertheless, even if made stronger, the second layer 50 still applies.

In a first modification, two different sized ceramic agglomerates, typically with suitable known binders, are treated in a manner similar to that used above for single sized agglomerates. , Mix evenly. Within this first configuration, there are two examples. In the first example, the agglomerate dimensions can be characterized as large and small grit, respectively, where the large agglomerates are about 0.053 to 0.028 inches (1.35 to 0.70 mm, Or 10 to 25 mesh) and the small agglomerates are about one-third to half of the size of the large grit, or about 0.030-0.012 inches (0.75-0.30 mm, or 20 to 50 mesh). It is noted that, although the mentioned ranges overlap, the large and small aggregates used are not expected to be of approximately the same size. Rather, it is contemplated that the differences in particle size will be distinct, with a ratio of at least about 1: 1.5, preferably about 2: 1. These two agglomerates of different sizes can be mixed in a weight ratio of small particles to large particles of approximately 1: 1 to 1:10. As can be imagined, the small agglomerates fill the void regions between the large coarse particles in the same manner as the small metal particles fill the interstices between the large agglomerates in FIG. Helps to reduce the size of the pathways through the layers of the, thus increasing the removal of solids by the filter.

As a second example, aggregate sizes can be characterized as large aggregates and very small aggregates. In this example, the large agglomerates may also be in the size range of about 0.053 to 0.028 inches (1.35 to 0.70 mm, or 10 to 25 mesh), but very small. Aggregates are about 0.0029-0. The dimensions are in the range of 0014 inches (0.075-0.035 mm, or 200-400 mesh). In this example, the two size agglomerates can be mixed in a weight ratio of very small particles to large particles of approximately 1: 2 to 1: 5.

In both of these examples, the smaller agglomerates allow the solid objects to at least partially fill the interstitial spaces between the larger agglomerates and reduce their size to pass through the filter. Reduce the size of things. That said, it can also reduce the amount of crossover between the first and second layers, as the smaller agglomerates can prevent the metal particles from entering the interstices. . Providing a suitable mixture of agglomerates of different sizes and applying the appropriate pressure can alleviate this problem. Moreover, such problems do not exist when the first and second layers are bonded together using the heat shrink method described above.

The second configuration for the first layer 48 is more layered, as aggregates of different sizes overlay the first and second layers in FIG. It is to overlap. In this configuration, aggregates of different sizes are not uniformly mixed and remain separate. Each of the different sized agglomerates is subjected to the above-described molding process to produce an untreated product, which is thin enough such that each untreated product forms a secondary layer of the finished filter. These layers, of course, overlap in the direction of gas flow through the filter.

Once the layers of greens have been made, they can be assembled together and placed in a mold, and pressed or pressed together to coalesce the greens. This creates a single layer as described above so that the amount of pressure applied to initially make the individual untreated secondary layers can be further compressed during this second step. This can be facilitated by reducing the number of steps compared to Although it is preferred to assemble all the secondary layers of the filter simultaneously in this second compression step, the individual layer or layers may be pressed and placed separately in place. This second pressing or compression step joins the different layers sufficiently to secure them together.

Pressing or pressing to form the monolayer 48 results in some mixing of aggregates of different sizes across the boundaries of adjacent layers. This mixing results in an interstitial transition zone of coarse grains of different sizes, similar to the homogeneously mixed different size aggregates discussed above. This transition zone is thus greatly reduced in voids and passages. As a result, the gas flow from the secondary layer of larger aggregates is diffused before entering the secondary layer of smaller aggregates downstream. This improves the filtering action as a more uniform flow is filtered by the secondary layer of smaller aggregates.

As would be expected, the final shape must fit within the inflator, where the thickness of the secondary layer depends on such final dimensions. However, the number of sublayers and their order can be varied to give the best results in a particular application.

The benefits of using multiple sized aggregates (blended or sub-layered) are not limited to the first layer of ceramic. In particular, the metal particles for the sintered second layer can be used essentially as described above. The only difference is that the metal particles can be made untreated by simply applying sufficient pressure if desired, and the temperature required to sinter the untreated material is known. It is possible to omit the binder as it will lower the sintering temperature. All other factors are essentially the same, including particle size and mixing ratio.

The final shape described above can also be obtained with a plasma arc metal layer. Specifically, two different particle sizes can be fed to the torch to provide a well-mixed layer. In order to make the layer overlap clearer, it is possible to grow (deposit) with particles of a first size and then grow further (deposit) with particles of a different size. As would be expected, particles grown by plasma arc acquire a texture similar to that of sintered metal particles.

From the foregoing, the present invention achieves all of the objects and objectives set forth above, as well as providing other advantages that are self-evident and inherent in the construction. Will be understood.

It will be appreciated that the given feature and sub-combination is of a working nature and can be used independently of other features and sub-combinations. This is expected and within the scope of utility model registration claims. Specifically, possible combinations of ceramic, metallic and non-metallic components are considered to be within the scope of this invention, even if not specifically mentioned for the sake of brevity. It is a thing.

Since many possible aspects can be constructed from the present invention without departing from the scope of the present invention, all the matters described herein or shown in the accompanying drawings are illustrated and interpreted. It should be understood that it should be done and not intended to be limiting.

[Brief description of the drawings]

FIG. 1 is a perspective view of a disc filter according to a first embodiment of the present invention.

2 is a cross-sectional view of a driver side inflator using the filter of FIG.

FIG. 3 is a perspective view of a disc filter according to a second embodiment of the present invention.

4 is a cross-sectional view of a driver side inflator using the filter of FIG.

FIG. 5 is a perspective view of a tubular filter according to still another aspect of the present invention.

6 is a cross-sectional view of a driver side inflator using the filter of FIG.

FIG. 7 is a perspective view of a passenger side inflator in which a filter according to the present invention forms a part of a housing.

FIG. 8 is an enlarged detailed view of an aggregate / particle structure according to the present invention.

FIG. 9 is an enlarged detailed view of a foam / particle structure according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Disc inflator 12 ... Upper surface 14 ... Lower surface 18 ... Airbag inflator 20 ... Upper wall 22 ... Bottom wall 24 ... Peripheral side wall 26 ... Inner wall 28 ... Inflator cavity 30 ... Gas outlet 32 ... Generator chamber 34 ... Pyrotechnic substance 36 ... Gas outlet 40 ... Tubular filter 42 ... Side wall 48 ... First layer 50 ... Second layer 52 ... Combined ceramic particles 54 ... Ceramic foam 56 ... Metal particles

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location B60R 21/26 B60R 21/26 (72) Creator Alexander Teverovsky USA, Massachusetts 01742, Concord, Dover Street 93 (72) Inventor George Sea. Mah-Janskiy United States, Utah 84405, Riverdale, West 4150 South 648

Claims (10)

[Utility model registration claims] 1. A strength sufficient to be used as a filter for gases generated by a pyrotechnic inflator,
A monolithic reticulated porous body having porosity and filter properties, the porous body being a first layer made of a porous ceramic, and fixed to each other and to the first layer. A unit for use in an inflator, comprising a second layer made from agglomerates of fixed metal particles, the second layer reinforcing the first layer and providing additional filtering action. Expression filter. 2. The first layer of porous ceramic is made from agglomerated ceramic particles or reticulated ceramic foam, and the metal particles of the second layer are sintered, mechanical interferenc.
The filter according to claim 1, which is fixed to the first layer by e) or plasma arc growth. 3. The filter of claim 2, wherein the first layer is an agglomerate of ceramic particles. 4. The filter of claim 2, wherein the first layer is a reticulated ceramic foam. 5. A method according to claim 2, wherein at least one of the first layer and the second layer is made up of first particles of larger size and second particles of significantly smaller size. Filter. 6. The filter of claim 5, wherein the second particles are small. 7. The filter of claim 6, wherein the first particles and the second particles are evenly distributed over at least one of the first and second layers. 8. The first particles form a first sublayer, the second particles form a second sublayer, and the first and second sublayers are directly 7. The filter of claim 6, wherein the filters are adjacent and the first particles and the second particles are intermingled at the intersection of these sublayers. 9. The first dimension is large and is 1.35.
The filter of claim 5, wherein the filter is in the range of 0.70 mm and the second dimension is very small, in the range of 0.075 to 0.035 mm. 10. The filter of claim 9, wherein the first and second particles are uniformly distributed throughout at least one of the first and second layers.